Biochemistry 1987, 26, 1194-1200
1194
Whitlow, M., & Teeter, M. M. (1985) J . Biomol. Struct. Dyn. 2, 831-848. Winterfeld, K., & Bijl, L. H. (1948) Justus Liebigs Ann. Chem. 516, 107-114.
Woodward, C., Simon, I., & Tiichsen, E. (1982) Mol. Cell. Biochem. 48, 135-160. Woynarowski, J. M., & Konopa, J. (1980) Hoppe-Seyler’s Z . Physiol. Chem. 362, 1535-1545.
Potato Tuber Cyclic-Nucleotide Phosphodiesterase: Selective Inactivation of Activity vs. Nucleoside Cyclic 3’,5’-Phosphates and Properties of the Native and Selectively Inactivated Enzymet Mdgorzata Zan-Kowalczewska, Jarodaw M. CieSla, Halina Sierakowska, and David Shugar* Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Rakowiecka 36, 02-532 Warszawa, Poland Received July 9, 1986; Revised Manuscript Received October 2, I986
ABSTRACT: Exposure of higher plant (potato tuber) cyclic-nucleotide phosphodiesterase (cPDase, EC 3.1.4) to alkaline p H a t 45 O C leads to irreversible selective loss of activity vs. 3’,5’-CAMP with full maintenance of activity vs. 2’,3’-CAMP. There is also loss of most (-80%) of the activity vs. the p-nitrophenyl esters of pT and T p but retention of full activity vs. p-nitrophenyl phenylphosphonate and bis(p-nitrophenyl) phosphate. The phosphonate ester, proposed as a specific substrate for 5’-nucleotide PDases [Kelly, S. J., Dardinger, D. E., & Butler, L. G. (1975) Biochemistry 14, 4983-49881, is now seen to be an excellent substrate for the activity of higher plant cPDase vs. nucleoside cyclic 2’,3’-phosphates. And, in fact, the phosphonate substrate competitively inhibits hydrolysis of 2’,3’-CAMP (Ki 0.3 mM) and vice versa (Ki 0.05 mM). Hydrolysis of the phosphonate ester is also inhibited by 3’,5’-cAMP, but with a Ki1-2 orders of magnitude higher. The potato enzyme, previously shown to be a tetramer, may be reversibly dissociated to the monomer, with full retention of all activities. Selective inactivation proceeds at the level of the monomer, following which reassociation to the tetramer is minimal and not accompanied by recovery of activity vs. 3’,5’-CAMP. Isoelectric focusing resolves the enzyme into five isozyme fractions, with p l values ranging from 6.8 to 8.1. All five fractions exhibit identical properties as regards molecular weight and substrate specifities both prior to and following selective inactivation. The foregoing findings readily account for conflicting reports on the properties of higher plant cyclic-nucleotide phosphodiesterases. Attention is drawn to some similarities with the nonconventional mammalian cyclic-nucleotide phosphodiesterase reported by Helfman and Kuo [Helfman, D. M., & Kuo, J. F. (1982) J. Biol. Chem. 257, 1044-10471.
-
-
Cyclic-nucleotide phosphodiesterase (cPDase)’ has been isolated and variously characterized from a variety of higher plants (Brown & Newton, 1981). Unlike the corresponding enzymes from mammalian cells, with one important exception (Helfman & Kuo, 1982; see below), the plant enzymes exhibit broad specificity and hydrolyze both cyclic 2’,3‘- and cyclic 3’,5’-phosphates of nucleosides (Vandepeute et al., 1973; Ashton et al., 1975; Shinshi et al., 1976; Zan-Kowalczewska et al., 1984). Rare exceptions include the enzyme from beans (Brown et al., 1977), reported inactive toward nucleoside cyclic 2’,3’-phosphates, and that from cultured tobacco cells, with very low activity vs. nucleoside cyclic 3’,5’-phosphates (Matsuzaki & Hashimoto, 1981), in disagreement with another report (Shinshi et al., 1976). This confusing situation has rendered difficult attempts to clarify the functional role of plant cPDases, the more so in that almost all preparations contain nucleotide pyrophosphatase activity (Zan-Kowalczewska et al., 1984). We have recently succeeded in purifying potato tuber cPDase to near homogeneity, with only residual (-0.5%) +This investigation profited from the support of the Polish Academy of Sciences (Project CPBR 3.13) and the Polish Cancer Research Program (Project CPBR 11.5).
0006-2960/87/0426- 1 194$01.50/0
pyrophosphatase activity (Zan-Kowalczewska et al., 1984). Apart from previously described activities, the purified enzyme also cleaved aryl esters of nucleoside 3’- and 5’-phosphates, nucleoside 5’-di- and 5’-triphosphates, aryl phosphonates, and the 2‘,3‘-cGMP terminal residue of a fragment of TMV RNA (Zan-Kowalczewska et al., 1984). During the course of further attempts to resolve the nature of the activity of this enzyme against such a multitude of substrates, it was observed that storage of the purified enzyme at 4 “C at pH 7.5 led to selective loss of activity against nucleoside cyclic 3’,5’-phosphates, without affecting activity vs. nucleoside cyclic 2’,3’-phosphates. This phenomenon has now been further examined, with results that undoubtedly contribute to a better understanding of the substrate specificity of this ubiquitous enzyme. I Abbreviations: PDase, phosphodiesterase; cPDase, cyclic-nucleotide phosphodiesterase; SDS, sodium dodecyl sulfate; SDS-PAGE, SDSpolyacrylamide gel electrophoresis; nitrophenyl-pT, thymidine 5‘-@nitrophenyl phosphate); Tp-nitrophenyl, thymidine 3’-(p-nitrophenyl phosphate); BIS, bis(p-nitrophenyl) phosphate; AS-BI-naphthol, 6bromo-2-hydroxy-3-naphthoicacid 2-methoxyanilide; AS-BI-naphthylpT, thymidine 5’-(AS-BI-naphthyl phosphate); NEM, N-ethylmaleimide; PMSF, phenylmethanesulfonyl fluoride; IEF, isoelectrofocusing; PEI, poly(ethy1enimine); Tris, tris(hydroxymethy1)aminomethane; EDTA, ethylenediaminetetraacetic acid.
0 1987 American Chemical Society
CYCLIC-NUCLEOTIDE PHOSPHODIESTERASE
EXPERIMENTAL PROCEDURES Materials. Sephadex IEF and LMW protein calibration kit were from Pharmacia (Uppsala, Sweden). Ampholines were from Serva (Heidelberg, FRG). Ampholine PAG plates, 1804-101 and 1804-103, were from LKB (Bromma, Sweden). Materials for electrophoresis and Triton X-100 were from BDH (Poole, Dorset, U.K.). Thymidine 5’-(AS-BI-naphthyl phosphate) was prepared as described elsewhere (Sierakowska et al., 1978). 2’,3’-[8-3H]cAMP (24.3 Ci/mmol) and 3’3’[8-3H]cAMP (26 Ci/mmol) were from Amersham (Bucks, U.K.), and p-nitrophenyl phenylphosphonate was from Regis Chemical Co. (Morton Grove, IL). NEM and PMSF were from Eastman Kodak (Rochester, NY). Other substrates and Fast Garnet GBC were from Sigma (St. Louis, MO). Thinlayer chromatography made use of Merck (Darmstadt, GFR) cellulose F and PEI-cellulose F plates. All other reagents were analytical-grade commercial products. Measurement of pH was carried out with a Radiometer (Copenhagen, Denmark) PHM 22p instrument. Enzyme. The highly purified, homogeneous cPDase previously described (Zan-Kowalczewska et al., 1984) exhibits only residual (-0.5%) pyrophosphatase activity vs. NADf. Because of its limited availability, due to the low yield of the purification procedure, most experiments were conducted with a preparation obtained following step 7 of the purification protocol and exhibiting 3-4% residual pyrophosphatase activity, referred to as partially purified cPDase. The findings with this preparation were checked periodically against those obtained with the highly purified enzyme and were essentially the same. In experiments involving incubation of the enzyme for more than several hours, a crystal of thymol was added to prevent possible bacterial contamination. Most experiments were based on the use of samples that, following selective inactivation, still posessed residual (0.4-2%) activity vs. 3’,5’-cAMP and, hence, also vs. AS-BI-naphthyl-pT (Zan-Kowalczewska et al., 1984). The latter substrate furnished a sensitive procedure for location of enzymatically active bands following gel electrophoresis (Bartkiewicz et al., 1985), as described below. Assays of enzyme activities were essentially as previously described (Zan-Kowalczewska et al., 1984), with incubation of an appropriate amount of enzyme with 5 mM substrate (unless otherwise indicated) in 0.05 mL of 0.1 M Tris-acetate buffer, pH 6, at 37 OC for 15 min. Under these conditions the reaction was linear with time and enzyme concentration. One unit of cPDase activity is the amount of enzyme hydrolyzing 1 pmol of 2’,3’-cAMP in 15 rnin at 37 OC. Kinetic studies, with 2’,3’-cAMP or 3),5’-cAMP as substrate, were performed radiochemically (Thompson et al., 1971). The 0.05-mL reaction mixture included 0.1 M Trisacetate buffer, pH 6.0,0.05-0.8 mM 2’,3’-cAMP plus 4 X lo4 cpm 2’,3’-[8-3H]cAMP (or 0.5-5.0 mM 3’,5’-cAMP plus 4 X lo4 cpm 3’,5’-[8-3H]cAMP), and appropriate levels of enzyme (and inhibitor), as indicated. Incubation was terminated by heating at 100 OC for 2 min. This was followed by addition of 0.15 mL of 0.1 M Tris-acetate buffer, pH 8.0, and 0.1 unit of alkaline phosphatase, incubation for 10 min at 37 “C, and then addition of 0.4 mL of water, followed by 1.O mL of an aqueous methanolic slurry of AG 1-X2 anion-exchange resin (Thompson et al., 1977). Liberated labeled adenosine was then measured as described by Thompson et al. (1971). With p-nitrophenyl phenylphosphonate as substrate (or inhibitor), the reaction mixture (0.75 mL) included 0.1 M Tris-acetate buffer, pH 6, 0.05-1.5 mM p-nitrophenyl phenylphosphonate, and appropriate levels of enzyme (and in-
VOL. 26, NO. 4,
1987
1195
hibitor) as indicated. Incubation, for 15 min at 37 “C, was terminated by addition of 0.05 mL of 1 N NaOH. Liberated p-nitrophenol was then measured as described by Razzell and Khorana (1961). Polyacrylamide gel electrophoresis (PAGE) was performed at pH 9.3 according to Davis (1964) on 0.4 mm thick 7.5% gel slabs. The enzyme solution was first concentrated by lyophilization and then dialyzed for several hours against 0.01 M Tris-acetate buffer, pH 6.0, containing 0.005% Triton x-100. SDS-PAGE was according to Laemmli (1970) additionally modified to permit detection of active forms of the enzyme (see below). The enzyme sample was treated for 2 h at 37 OC with 2.3% SDS in 0.0625 M Tris-HC1 buffer, pH 6.8, and subjected to electrophoresis on 0.4 mm thick 10.5% gel slabs at 4 OC,initially at 10 mA for 15 rnin and subsequently at 20 mA for 3 h. Marker proteins were treated with SDS (Laemmli et al., 1970), electrophoresed as above, and stained with Coomassie blue R-250. Localization of cPDase activity on gels (and also following IEF) was as follows. The gel was rinsed for 5-10 min twice with cold 0.2 M acetate buffer, pH 5.2, and then either reacted histochemically for cPDase activity or sectioned into 5-mm lengths and assayed against several substrates. Histochemical reactions were performed by incubation of a gel strip in 12 mL of 0.1 M acetate buffer, pH 5.2, plus 10 mM EDTA with 1 mg of thymidine 5’-(AS-BI-naphthyl phosphate) and 1.5 mg of Fast Garnet GBC at 37 “C for up to 2 h (Sierakowska et al., 1978). Enzyme Activity of Gel Fractions. With nitrophenyl-pT or BIS as substrates, the gel fraction was incubated in 0.1 mL of 5 mM solutions, as above. With 2’,3’-cAMP and 3 ’ 3 CAMP as substrates, concentrations were 5 mM in 0.2 mL, and the reaction was terminated by heating for 3 min in boiling water, followed by addition of 0.05 mL of 0.8 M Tris-HC1 buffer, pH 8.0, plus 0.1 unit of alkaline phosphatase and incubation for 10 rnin at 37 OC. For color development, 0.2 mL of the medium overlaying the gel was withdrawn and the enzymatically liberated Pi assayed as described elsewhere (Bartkiewicz et al., 1985). Zsoelectrofocusing (IEF), preparative and analytical, was carried out with an LKB 21 17 flat-bed apparatus (Multiphor), as described in LKB Application Note No. 198 (preparative) and No. 250 (analytical). The enzyme sample was first concentrated by lyophilization and dialyzed vs. 0.1 M glycine with 0.005% Triton. Preparative IEF was with 3-mm Sephadex IEF and ampholines at pH values of 6.0-8.5 for 16 h at 4 “C, initially at 560 V and 14 mA and subsequently at 1900 V and 7 mA. A paper print was then taken of the gel solution (30 s) to reveal the focused protein zones. The print was dried immediately and sprayed with a solution for histochemical localization of activities. Relevant bands were removed from the gel and eluted with 0.1 M acetate buffer, pH 5.2. The enzyme was salted out by addition of (NH,),SO4 to 60% saturation; the protein precipitate was collected on a Millipore 0.45-pm filter and dissolved in 0.1 M Tris-HC1 buffer, pH 7.5, containing 0.02% Triton. The individual fractions were dialyzed for 2 h vs. 0.05 M Tris-acetate buffer, pH 6.0, containing 0.01% Triton and used in this form for further analysis. The pH along the gel was determined by cutting out 2 cm side strips 0.5 cm in length. Each strip was eluted into 1 mL of water and the pH determined with a Radiometer 22p instrument with a semimicro glass electrode. Sucrose density gradient centrifugation was performed as previously described (Zan-Kowalczewska et al., 1984).
1196
BIOCHEMISTRY
ZAN-KOWALCZEWSKA
ET AL.
Table I: Changes in Relative Rates of Hydrolysis of Various Substrates of Potato Tuber cPDase, following 90% Selective Inactivation of Activity vs. 3’,5’-CAMPby Heating at 45 “C, pH 8.8, for 18 h“
61
I
native enzyme
substrate
selectively inactivated enzymeb
rate (5%) K , (mM) rate (%) K , (mM)
2’,3’-cAMP
100
0.4
100
3‘,5‘-cAMP bis@-nitrophenyl)
23 150
1.5
2 105c
0.4
0.4 I .5 0.4
phosphate
‘i,
‘t ‘
4
u 1 4
5
6
8
7
9
10
PH FIGURE 1: Influence of pH on activities of purified potato tuber cPDase and 2’,3’-CAMP (0),following incubation at 45 vs. 3’,5’-CAMP (0) O C for 18 h in 0.1 M buffers (acetate, pH 4.5-6; Tris-HCI, pH 7-9; glycine, pH 10). Controls were identical samples at each pH stored at -20 OC. Activities are in micromoles of substrate hydrolyzed in 15 min at 37 O C . Identical results were obtained with cytidine or deoxycytidine cyclic phosphates.
Hydrolysis products of 2‘,3‘-CAMP were determined chromatograpically on Merck cellulose F plates, as described by Helfman and Kuo (1982), and of 3’,5’-cAMP with the use of the 3H-labeled compound on Merck PEI-cellulose F plates according to Ashton and Polya (1975). Selective Inactivation of Activity us. 3’,5‘-cAMP. Optimal conditions were as follows: the enzyme sample was dialyzed vs. 0.1 M Tris-HC1 buffer, p H 8.8, containing 0.01% Triton for 2 h at 4 O C . Enzyme activities were monitored, and the solution was divided into two aliquots. One was stored at -18 O C as a control; the other was kept a t 45 O C for periods of up to 18-30 h. An additional control consisted of a sample dialyzed vs. 0.1 M Tris-acetate buffer, pH 6.0, and then kept at 45 O C .
RESULTS The starting point for this study was the unusual observation that storage a t 4 OC, in 0.02 M Tris-HC1 buffer, p H 7.5, of a preparation of potato tuber cPDase established as homogeneous by various criteria (Zan-Kowalczewska et al., 1984) led to a slow time-dependent decrease in activity vs. 3 ‘ 3 cAMP (and nitrophenyl-pT, see below) with full maintenance of activity toward 2’,3’-cAMP. Selective Inactivation of cPDase. Storage of the enzyme solution at p H values
